Many of the agents commonly used for cancer therapy kill tumor cells by damaging cellular DNA, and thus DNA repair proteins and their interactions represent potential targets for enhancing such therapy. Double- strand breaks (DSBs) in DNA, which are the major cytotoxic lesions formed by ionizing radiation as well as by topoisomerase II inhibitors, are repaired primarily by nonhomologous end joining (NHEJ) and, to a lesser extent, by homologous recombination (HRR). Thus, the efficiency of these repair systems is a critical factor in the effectiveness of radio- and chemotherapy. XLF is the most recently discovered core protein in the NHEJ pathway. Although genetic studies show that XLF is essential for efficient NHEJ, its precise role in the pathway is poorly understood. In addition, because the interaction between XLF and another NHEJ protein, XRCC4, is critical for NHEJ, but is transient and has only moderate (low ?M) affinity, it represents an attractive target for chemo/radiosensitization by disruption of NHEJ. The primary objectives of this application are, therefore, to elucidate in more detail the role of XLF in NHEJ, and to develop XLF/XRCC4 interaction inhibitors and evaluate them as chemo/radiosensitizers. In order to better clarify XLF's role and test the hypothesis that XLF and XRCC4 polymerize into a filament that aligns DNA ends, photoactivatable DNA substrates will be added to cell extracts to map the interactions of XLF with DNA near a DSB as a function of time and distance from the break. Experiments with a fluorescent NHEJ reporter integrated into the genome will determine whether alignment-based gap filling at DNA ends is strictly dependent on XLF in intact cells, as has been already demonstrated in cell extracts. For the inhibitor development, the novel in vitro selection technique, mRNA display, will be used to isolate small (Mr 1000-2000), cell-penetrating, drug-like cyclic peptides that bind to XRCC4 at its interface with XLF. Using mRNA display, an extremely diverse library of 10 trillion semi- random peptides fused to the mRNAs that encode them will be prepared using in vitro translation. Those that bind to the XLF binding pocket on XRCC4 will be selectively captured and their mRNA amplified with RT-PCR. After several rounds of iterative translation, selection, and amplification, candidate peptides that bloc the XLF/XRCC4 interaction will be identified. These peptides will be screened for NHEJ inhibition using a cell extract-based assay wherein joining of partially complementary DSB ends is stringently dependent on XLF. The strongest inhibitory peptides will be chemically synthesized with addition of cell-penetrating tags, and tested for sensitization of triple-negative breast tumor cells to two DSB-inducing agents: ionizing radiation and etoposide. The most potent sensitizers will be used in further experiments to determine to what extent DSB repair is inhibited in several cell lines, chosen either for their known radioresistance, or because they are deficient in DSB repair by the alternate HRR pathway. Based on their unique mode of action, these peptides or subsequent derivatives thereof may be clinically useful in sensitizing tumor cells for radio- and chemotherapy.